Non-ST elevation acute coronary syndrome (NSTE ACS) is a both common and mortal condition in the elderly. Acute antithrombotic therapy is efficacious, yet under-utilized in older patients due in part to their associated bleeding risks. Importantly, we have new evidence that these risks are likely modifiable via individualized antithrombotic drug dosing. In Specific Aim 1 we propose to use detailed clinical information from CRUSADE, the nation's largest NSTE ACS registry (with 450+ existing hospitals, and an additional 130+ Veterans Health Administration (VHA hospitals) to determine the incidence of bleeding in elderly NSTE ACS patients as well as the clinical predictors of this risk. In Specific Aim 2, we will determine specific modifiable care factors associated with bleeding risks. We will assess the frequency that antithrombotic dosing guidelines are exceeded in current practice, the generalizability of these rules in elderly patients, as well as the consequences associated with excessive dosing. We will then conduct a multicenter randomized control trial to reduce NSTE ACS bleeding events (Specific Aim 3). The primary goals of our quality improvement intervention will be to reduce bleeding rates in the elderly via increasing adherence to antithrombotic dosing guidelines. Improved drug safety should also increase the overall appropriate use of antithrombotic agents in elderly NSTE ACS patients, and ultimately, to improve patient outcomes. We also will critically evaluate factors associated with the success of our quality improvement intervention; and ascertain whether such efforts need to be maintained to sustain improved performance (Specific Aim 4). Finally, we will use established relations with professional societies and the VHA to widely disseminate our study's major findings.